Baghdad99/saad-hausa-text-to-english-text

Baghdad99/saad-opus-ha-to-en

This model is a fine-tuned version of Helsinki-NLP/opus-mt-ha-en on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 1.9691
• Validation Loss: 3.9168
• Epoch: 24

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Epoch
5.3200	4.9969	0
4.7301	4.7199	1
4.3658	4.5332	2
4.0897	4.3940	3
3.8568	4.2969	4
3.6818	4.2143	5
3.5195	4.1530	6
3.3799	4.1085	7
3.2595	4.0679	8
3.1400	4.0438	9
3.0337	4.0185	10
2.9335	3.9988	11
2.8354	3.9766	12
2.7461	3.9684	13
2.6650	3.9585	14
2.5827	3.9477	15
2.5062	3.9422	16
2.4289	3.9306	17
2.3549	3.9208	18
2.2882	3.9205	19
2.2195	3.9183	20
2.1492	3.9157	21
2.0888	3.9163	22
2.0302	3.9160	23
1.9691	3.9168	24

Framework versions

• Transformers 4.35.0
• TensorFlow 2.14.0
• Datasets 2.14.6
• Tokenizers 0.14.1